Urban security management is inseparable from the camera. Currently, a large number of identification data such as license plates, faces and features cannot be analyzed manually. A single 1080P camera can generate about 80G of data per day, if these are traditionally Data is passed back to the cloud server for video analysis, which will generate a large amount of transmission and storage resources. At the same time, the truly valuable data in these massive data is far less than 15%. In order to reduce costs and improve efficiency, the transition from traditional centralized big data processing to edge big data processing is the trend of the times. A new technology for cloud-edge integration was born.
The biggest difference between edge computing and cloud computing is that data will no longer need to be transmitted to distant clouds, and data processing can be solved on the edge side. This also solves the uniqueness of the cloud on the data to achieve smart applications. This approach brings an unprecedented experience:
From a cost perspective, bandwidth and storage costs can be saved. IDC statistics show that cloud computing and edge computing combine to cost only 39% of cloud computing alone; from an efficiency perspective, because edge computing is closer to users, The edge node realizes the filtering and analysis of the data. The massive device data does not need to be uploaded to the cloud and is fed back to the terminal for execution. From the perspective of the integrated service level, cloud edge fusion can greatly enhance the deep learning ability of artificial intelligence, AI+ edge. The combination of calculations allows edge calculations to be more than computational, and intelligent functions such as target detection, feature extraction, and lossless modeling can be achieved.
Hikvision believes that the integration of edge computing and cloud computing can better solve the real problem of the Internet of Things. Taking video surveillance as an example, it is necessary to store the original video stream and perform intelligent analysis. It also needs to realize on-demand aggregation and big data analysis of structured data. It also needs unified management and operation and maintenance of various Internet of Things and IT equipment. . Therefore, an intelligent architecture that can achieve edge computing and centralized processing of massive heterogeneous data on demand is needed to realize the organic integration of artificial intelligence, big data, cloud computing and terminal equipment. In this context, Hikvision has proposed AI Cloud based on the cloud-edge fusion concept.
Artificial intelligence needs to build AI ecologyArtificial intelligence is a systematic project. For enterprises, whether they want to empower or promote industrial development, the construction of an ecological system is indispensable. At present, the industrial chain of artificial intelligence involves different links from basic algorithm models, basic hardware, basic products, industrial applications, scenario applications, solutions to application delivery. No single enterprise can independently create a complete industrial chain, and work together. Solving the needs of users is the trend of the future era of artificial intelligence.
Hikvision President Hu Yangzhong said at the "Zhiyong Qiantang" summit: "At present, the application of artificial intelligence is scene-based, and the application of scenes is fragmented. It is difficult for fragmented applications to concentrate on forming fists. Because of the long chain of intelligent industry, there are many aspects involved, and it is necessary for the related parties in the industry to cooperate with each other. In today's global economic integration, global artificial intelligence enterprises need to cooperate to jointly promote the development of artificial intelligence industry and benefit. Human society."
For the long industrial chain of fragmentation market and artificial intelligence, to create industrial innovation and sustainable development of industrial ecology, algorithms, computing power, products, data, training systems, software platforms, application software and other parties need to participate. Therefore, Hikvision adheres to the principle of “openness, cooperation and win-winâ€. It is building a complete system of open capabilities through the opening of infrastructure, data resources, platform services and application interfaces, and collaborating with partners to build a new ecosystem of AI Cloud. At the same time, it promotes industry solutions based on AI Cloud in various industries, fosters and develops professional partners in sub-sectors and sub-sectors, and builds an industry eco-cooperation system. It is understood that Hikvision will continue to pay attention to the development of the industry chain, such as consulting, operation and maintenance, security protection and other service ecology.
It can be seen that Hikvision's AI Cloud is not at all simple, not only the development concept, technical architecture, but also product realization, service mode, and open ecology.
AI Cloud is born for industrial intelligence upgradeAt the “Zhiyong Qiantang†summit, Hikvision gave a detailed account of the value of AI Cloud.
Bihui Juan, senior vice president of Hikvision, emphasized that in AI Cloud, edge nodes focus on multi-dimensional sensing data collection and front-end intelligent processing; edge domains focus on sensing data aggregation, storage, processing and intelligent applications; cloud centers focus on IoT data. The fusion of multidimensional data and multidimensional analysis applications based on big data. Data from the edge node to the edge domain, to achieve "gather edge to domain"; from the edge domain to the cloud center, to achieve "data into the cloud." Domains and cloud centers can be multi-level and multi-class. Depending on the application, the data aggregated in the edge domain and the data transmitted to the cloud will be different in model and content. The edge domain acts like a “midfield†and is responsible for deciding when, what type, and to what extent data is sent to the cloud center for “on-demand aggregationâ€.
From the above point of view, you can understand that the cloud-edge fusion of Hikvision AI Cloud is not a simple cloud + edge, but it is not simply divided into several so-called "small clouds" and then aggregated into a larger cloud, but it is solved A series of issues such as application, data processing, and management brought by the edge to the cloud. In short, AI Cloud is not Cloud CompuTIng, but the sum of infrastructure, data resources, platform services, and applications.
Xu Hicming, senior vice president of Hikvision, added that both the public security and commercial areas need to better sense the front end through video or other Internet of Things to achieve rapid response and agile disposal of the entire grid.
According to the author's understanding, based on AI Cloud, Hikvision has been oriented to the public security, transportation, justice, culture and education, energy, finance, intelligent buildings and other industries, and strive to create AI resources schedulable, data on-demand convergence, scene application And four integrated capabilities such as integrated operation and maintenance services, enabling intelligent transformation of various industries. For example, in the field of public safety, AI Cloud helps fight crime and protect everyone's safety. In the public service field, AI Cloud provides smart travel and creates a green and harmonious ecological environment. In the business field, AI Cloud helps owners optimize products and services. Operating efficiency; In the financial sector, AI Cloud empowers precision marketing and enhances customer experience; in education, AI Cloud will enable smart education, innovative teaching management and more.
The value of AI CloudThe currently deployed cameras in China generate 66 terabytes of data per second. If the data is transmitted only by the cloud data center, the transmission cost and storage cost will remain high, but if it can be stored and processed on the edge of the network, That will greatly reduce costs and improve equipment processing efficiency.
The development of smart security requires not only the cloud computing of "high on" but also the edge calculation of "grounded gas". On the one hand, edge computing collects data information, performs preprocessing and extracts feature data and transmits it to the cloud brain. On the other hand, it helps to better implement IoT applications by opening up various system platforms. Regardless of the efficiency of use of IoT applications, time delays, or security considerations, edge computing is the key to the popularity of the Internet of Things.
This is also the value of Hikvision's AI Cloud. From the microscopic point of view, it is the dominant goal to process the data that does not need to go up to the cloud. From a macro perspective, Hikvision and Huawei are in the same way in the ICT industry. It is striving to assume the role of “leading big brother†in the industry and promote the construction of a good security ecology. The open platform, open experience environment and compatibility verification environment are laying a good foundation for “Security + AI†and also a partner. Build a stage of cooperation and win-win, and promote the application of AI.
It can be said that the "security elephant" is dancing in a new attitude.
1.27mm Pitch
1.27mm Pitch
ATKCONN ELECTRONICS CO., LTD , https://www.atkconn.com